The sensorimotor control of human posture will be investigated theoretically, in close collaboration with experimental studies and clinical programs. The proposed studies continue ongoing explorations of simple ways the nervous system can control posture by stored, habitual responses, using little on-line calculation. This proposal is divided into three major projects: The first study will be on postural mechanics as controlled by muscle burst patterns or synergies. We have described the structure and mechanics of these synergies and their mixtures in the quasi-static case for two dimensions. This study will expand the analysis to non-static cases and to a third dimension, allowing the analysis to be applied to more patient groups and more natural situations. The mathematical approach will allow a generalization to other movements, such as arm and head movements, and a statement of the types of movement trajectories that can be controlled by analogous combinations of muscle burst patterns. The second study will describe the coordination of orientation senses, visual, vestibular, and tactile, in controlling movements. Sensory feedback from the three different orientation senses must be properly blended to give an accurate representation of movement. We will investigate how this is done by both normal subjects and groups of vestibular patients, and how sensory deficits in vestibular patients affect their movements. The third project is to find constraints on a combined sensorimotor control system which allows both proper interpretation of sensory inputs and appropriate mechanical interaction of muscle forces with body mechanics and the support surface. This project will use the results of the previous two.